Leveltree 1

I’ve been working on the progression system for an RPG project I’m developing, and I’d like to talk about it a little bit.

In a game where the player is able to choose how to develop their units, it’s difficult to create a situation where all the builds the player could make are around equal in effectiveness. Usually there will be one that is stronger than the others. In games like these, part of the fun to crack the system and figuring out what that optimal build is, but that’s only something that can be done once, since after you know what that best build is, you can just go for it every time. Of course, as a player, you can choose to make a worse build either to challenge yourself or for variety’s sake or any other reason, and this fine—it’s something I do frequently myself—but I do think this kind of situation can be improved upon.

For my system, I’ve decided to focus on a few things that I enjoy about progression systems and to build the system around that. Those are:

Tradeoffs: Every time the player goes to learn a new move, they have to chose between two of them. They’ll only get to pick one.

Flexibility: I’ve never liked things like talent trees. They force you to make an early decision about the type of character you’re building. Early decisions are less desirable, I think, since you have less information at that point. And having the chosen talent tree decide what type of character you’re building further limits the character you can make. Essentially, many talent tree systems seem to condense the character type question down into one early decision. Keeping a flexible system allows the player to build their characters in response to weaknesses they notice while playing.

For my system, every time the player levels up, the player chooses between two moves to learn. One they get to keep, and the other they lose for good. Since each of the player’s units can have 5 moves equipped, and can know something like 13 moves in total, each decision still has a low overall impact on their build. In fact, by swapping the equipped moves, a unit can have its role tweaked between missions. It separates each decision from the others too, which lowers the stakes, while still requiring the player to make a choice, but only a small one. Small decisions are easier to make and there’s less of a penalty for changing your mind. Finally, breaking down the system to a move-by-move level means that there is a much higher number of total possible builds in the end.

Units learn moves in the above way, and they have their stats increased in a similar fashion. Namely, each move they learn has an associated set of stat increases with it, and when the player learns a move, they have their stats increased by the accompanying values. This adds some additional depth to the move learning decision, since different moves will increase stats by different amounts.

I never thought I had created a perfect, or even good, system, and I certainly don’t think there is one singular good approach for a progression system. But having written this, I think I have created, at the very least I think, a system that the player will have to interact with. And that’s what a game is, really. Hopefully it’s fun too.